A mouse finds itself in a box it’s seen before; inside, its white walls are bright and clean. Then, a door opens. On the other side, a dark chamber awaits. The mouse should be afraid. Stepping into the shadows means certain shock — 50 hertz to the paws, a zap the animal was unfortunate enough to have experienced just the day before. But when the door slides open this time, there is no freezing, no added caution. The mouse walks right in.
The memory of this place, of this shock, of these bad feelings had been erased overnight by a team of neuroscientists at four leading research institutions in Japan using lasers, a virus, and a fluorescent protein normally produced in the body of sea anemones. Their work, published Thursday in Science, pinpoints for the first time the precise timing and location of minute brain changes that underlie the formation and consolidation of new memories.
“It’s a huge stride for the field,” said Steve Ramirez, a neuroscientist at Boston University who was not involved in the study.
That field is optogenetics, a new lab technique that makes neurons sensitive to light. And it has allowed scientists to wield lasers like memory-editing wands, launching a new era of exploration into the inner workings of animals’ brains. Because nearly every neurological disease — from depression to PTSD to Alzheimer’s — impacts the memory system, discoveries in animals could someday lead to new treatments for these disorders in people. (In one small but promising study, optogenetics helped a blind man see for the first time in 40 years.)
Scientists like Ramirez have been able to use these techniques to implant mice with false memories and crank the emotional dial of real ones. But he and others achieved such feats of optogenetic inception by lighting up whole brain areas. “We’re kind of hitting the cells with a sledgehammer because we’re activating the cell body, the axons, everything,” Ramirez said. In contrast, the research team from Japan developed a way to toggle not just specific neurons but specific synapses on those neurons — the exact places where they connect with other neurons — during specific time windows. “That extra layer of resolution is technically impressive and particularly exciting because it means we’re getting close to finding a causal mechanism for memory,” he said.
Contrary to what Hollywood would have you believe, memories exist as a three-dimensional phenomenon in the brain. You can’t go to one single X, Y, Z coordinate point and say “this is where a memory lives” (and if that memory is painful, cut it out, a la “Eternal Sunshine of the Spotless Mind”). Rather, memories reside across different parts of the brain, grooved into networks of neighboring neurons like webs of interleaved dominos. Knock any one of them over and the cascade that follows might lock a new memory in or conjure up an old one.
One way that scientists think that happens is through something called long-term potentiation. All the sights and sounds and smells and emotions associated with a given experience cause certain neurons to fire. And when they do, it leads to enduring changes in those cells and in cells nearby — they sprout protrusions that help transmit electrical signals and make more connections with nearby neurons. The strengthening of activity within and between brain cells is believed to be a substrate for memory.
Nearly two decades ago, Yasunori Hayashi, a neuroscientist then at the RIKEN Center for Brain Science, discovered some of the critical molecules involved in the physical changes taking place during the early phases of long-term potentiation. More recently, his lab figured out how these key proteins rapidly rearrange and polymerize to produce these protrusions. In this latest study, Hayashi, now at Kyoto University Graduate School of Medicine, led work developing a method to selectively disrupt that process using light. He and his collaborators used a viral vector to make neurons express SuperNova, a sea anemone protein that releases destructive oxygen species when illuminated. Shine a light on them and they paralyze their surrounding proteins — no protrusions, no long-term potentiation.
“This technology using light made it possible for the first time to erase long-term potentiation only at a specific time window,” Akihiro Goto, the study’s lead author, told STAT via email. He and his colleagues used this technology to disrupt the long-term potentiation of neurons in the hippocampus — a brain region crucial for memory formation — at regular intervals in the hours and days following the shock-box training. They discovered that in order for memories to form, long-term potentiation was required immediately after the event and in the subsequent sleep period.
To see if they could erase memories both coming and going, the researchers also blocked long-term potentiation in the anterior cingulate cortex, a different brain region involved in recalling more distant memories. When blocked two days out from the training, poof, the memories disappeared. But when they tried it on day 25, the animals found their fear. By then, the memory had stuck.
“Our study revealed a precise spatiotemporal profile of long-term potentiation in hippocampus and the anterior cingulate cortex, and demonstrated that such local long-term potentiation in each brain region is required for the early phase of memory consolidation,” wrote Goto, adding that it also explains why sleep is important for memory formation “and is expected to contribute to the treatment of sleeping disorders.”
That’s not to say that doctors are going to be sticking optic fibers into the human brain anytime soon. But Ramirez sees the work as a positive step toward developing different interventions, such as drugs, that could erase certain fearful memories like those involved in PTSD. “That specificity is important because it means that in theory it should be possible to develop drugs that only reverse this kind of plasticity in the brain, making it so we can erase or suppress only very specific memories,” said Ramirez. The downside is that long-term potentiation is not the only mechanism by which our brains store information. So more research is needed to understand what those different mechanisms are before optogenetics can be used to illuminate which kinds of memories they encode.
“It’s still sci-fi sounding,” said Ramirez. “But it doesn’t require inventing a new law of physics.”